Menu
Analyze the full ecosystem and surface what needs doing next — including emergent tasks not on the board
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Scan all ecosystem repos for convention deviations. Reports findings; does not fix anything. Use proactively after cross-repo changes.
Validates pipeline parquet outputs for schema compliance, null rates, row count baselines, and freshness. Use after pipeline runs or when triage shows data quality concerns. Fast and read-only (uses haiku).
Run tests across all ecosystem repos and report pass/fail summary. Use when the user wants to verify the ecosystem is healthy or after cross-repo changes.
Runs the krff-shell ETL pipeline with pre/post validation. Use when data needs refreshing, after dependency updates, or when triage shows stale parquets. Reads CLAUDE.md before running.
Research across sibling repos to gather context for cross-project tasks. Use when working in one repo and needing to understand how another repo works, what it exports, or how data flows between them.
Show all open work (board + backlog) with dependency-aware execution order
| name | plan |
| description | Analyze the full ecosystem and surface what needs doing next — including emergent tasks not on the board |
| user-invocable | true |
| disable-model-invocation | true |
| allowed-tools | Bash, Read, Grep, Glob, Agent |
Analyze the ecosystem and produce an actionable plan. This is a read-only operation — do NOT fix anything, do NOT add board items, do NOT modify files.
conventions: Run only the convention audit layer with full detailconventions:Spawn the convention-auditor agent and wait for its result. Display the full convention audit table. Done.
Run these 5 layers, parallelizing where possible:
Layer 1 — Board State (parallel with 2 and 3)
Fetch the board:
"/c/Program Files/GitHub CLI/gh.exe" project item-list 1 --owner pon00050 --format json
Check for:
Layer 2 — Repo Health (parallel with 1 and 3)
For each repo in C:\Users\pon00\Projects\ (krff-shell, kr-forensic-core, kr-dart-pipeline, kr-anomaly-scoring, kr-stat-tests, kr-company-registry, kr-trading-calendar, kr-beneish, kr-derivatives, jfia-catalog, jfia-forensic, kr-enforcement-cases, kr-real-estate):
git -C ../REPO status --porcelain
git -C ../REPO log --oneline -1 --format="%cr"
git -C ../REPO rev-list --count @{u}..HEAD 2>/dev/null # unpushed commits
Flag:
## Known Gaps section in CLAUDE.md (convention drift)Layer 3 — Convention Drift (parallel with 1 and 2, via background agent)
Spawn the convention-auditor agent in the background:
Use Agent tool with subagent_type convention-auditor, run_in_background: true
Continue with layers 1-2 while this runs. Incorporate results at the end.
Layer 4 — Integration Gaps (after layers 1-2)
Check across repos:
tests/ but no conftest.pysrc/ but no _paths.py or paths.pyUse Glob and Grep across C:\Users\pon00\Projects\*.
Layer 5 — Strategic Alignment (last, synthesizes everything)
Read ECOSYSTEM.md in the hub. Cross-reference with board state from Layer 1:
=== Ecosystem Plan ===
Board: X todo (Y AI, Z human), W done
Health: X/8 repos clean, Y need attention
Emergent Tasks (not on board):
1. [HIGH] description — repo — rationale
2. [MED] description — repo — rationale
...
Suggested Next Session:
AI: [specific task] — [why now]
Human: [specific task] — [deadline if any]
Convention Drift: X deviations found (Y since last audit)
[summary table from convention-auditor]